Piecewise-constant parametric approximations for survival learning

Jeremy C. Weiss
Proceedings of the 2nd Machine Learning for Healthcare Conference, PMLR 68:1-12, 2017.

Abstract

Logged events occur both regularly and irregularly over time. In electronic health records, these events represent mixtures of scheduled and urgent or emergent encounters. Whereas most survival models use baseline events to estimate the rate function for an outcome, e.g., Cox processes using the proportional-hazards assumption, our framework uses logged events over time to predict survival outcomes with piecewise approximations of arbitrary hazard functions. We develop a procedure to learn forests as combinations of piecewise-constant and parameterized distributions to compactly model survival distributions from data. Under this construction, the model provides a “now-time” risk that incorporates irregularly-repeated data and for health outcomes serves as a surrogate for patient disposition. We illustrate the advantages of our method in simulations and in longitudinal, intensive care unit data of individuals with diabetes admitted for ketoacidosis.

Cite this Paper


BibTeX
@InProceedings{pmlr-v68-weiss17a, title = {Piecewise-constant parametric approximations for survival learning}, author = {Weiss, Jeremy C.}, booktitle = {Proceedings of the 2nd Machine Learning for Healthcare Conference}, pages = {1--12}, year = {2017}, editor = {Doshi-Velez, Finale and Fackler, Jim and Kale, David and Ranganath, Rajesh and Wallace, Byron and Wiens, Jenna}, volume = {68}, series = {Proceedings of Machine Learning Research}, month = {18--19 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v68/weiss17a/weiss17a.pdf}, url = {https://proceedings.mlr.press/v68/weiss17a.html}, abstract = {Logged events occur both regularly and irregularly over time. In electronic health records, these events represent mixtures of scheduled and urgent or emergent encounters. Whereas most survival models use baseline events to estimate the rate function for an outcome, e.g., Cox processes using the proportional-hazards assumption, our framework uses logged events over time to predict survival outcomes with piecewise approximations of arbitrary hazard functions. We develop a procedure to learn forests as combinations of piecewise-constant and parameterized distributions to compactly model survival distributions from data. Under this construction, the model provides a “now-time” risk that incorporates irregularly-repeated data and for health outcomes serves as a surrogate for patient disposition. We illustrate the advantages of our method in simulations and in longitudinal, intensive care unit data of individuals with diabetes admitted for ketoacidosis.} }
Endnote
%0 Conference Paper %T Piecewise-constant parametric approximations for survival learning %A Jeremy C. Weiss %B Proceedings of the 2nd Machine Learning for Healthcare Conference %C Proceedings of Machine Learning Research %D 2017 %E Finale Doshi-Velez %E Jim Fackler %E David Kale %E Rajesh Ranganath %E Byron Wallace %E Jenna Wiens %F pmlr-v68-weiss17a %I PMLR %P 1--12 %U https://proceedings.mlr.press/v68/weiss17a.html %V 68 %X Logged events occur both regularly and irregularly over time. In electronic health records, these events represent mixtures of scheduled and urgent or emergent encounters. Whereas most survival models use baseline events to estimate the rate function for an outcome, e.g., Cox processes using the proportional-hazards assumption, our framework uses logged events over time to predict survival outcomes with piecewise approximations of arbitrary hazard functions. We develop a procedure to learn forests as combinations of piecewise-constant and parameterized distributions to compactly model survival distributions from data. Under this construction, the model provides a “now-time” risk that incorporates irregularly-repeated data and for health outcomes serves as a surrogate for patient disposition. We illustrate the advantages of our method in simulations and in longitudinal, intensive care unit data of individuals with diabetes admitted for ketoacidosis.
APA
Weiss, J.C.. (2017). Piecewise-constant parametric approximations for survival learning. Proceedings of the 2nd Machine Learning for Healthcare Conference, in Proceedings of Machine Learning Research 68:1-12 Available from https://proceedings.mlr.press/v68/weiss17a.html.

Related Material